Skip to main content
. 2022 Sep 25;5:100287. doi: 10.1016/j.fsisyn.2022.100287

Table 3.

Number of classes obtained by CHD per corpus and distinction of a class related to false documents. N indicates the number of analysed ads for every corpus. The first ten words of each class found are also reported.

Corpus Number of classes Top 10 words in each class (by number of occurrences)
Fakedoc_onlinebusiness (N = 9813) 4 account; warranty; premium; porn; lifetime; extra; market; cheap; bonus; month
hq; psd; template; card; id; scan; license; driver; passport; dl
hq; usa; card; bank; cc; fullz; fresh; balance; email; verify
database; record; hack; leacked; plaintext; million; dtabase; leak; voter; log
Fakedoc_forgeries (N = 2812) 4 psd; template; id; driver; license; passport; scan; hq; card; statement
replica; perfect; shoe; vuitton; louis; lv; gucci; black; bag; dior
series; gold; black; watch; rolex; box; pro; counterfeit; clone; max
fakemoney; series; eur; test; pen; pass; uv; usd; version; stripe
Fakedoc_software (N = 3489) 4 pro; full; crack; program; macos; adobe; x64; window; pack; hack
full; software; mac; source; tool; code; bitcoin; rat; android; stealer
premium; porn; video; account; lifetime; movie; book; private; spotify; proxifier
psd; template; id; driver; license; passport; scan; card; hq; statement
Fakedoc_services (N = 3436) 6 psd; template; id; driver; license; passport; statement; fake; utility; usa
id; scan; passport; utility; custom; usa; quality; dl; high; bill
complet; credit; full; uk; pack; list; delivery; real; utter; service
card; hq; egift; pdf; restaurant; grill; pizza; italian; group; bar
account; lifetime; premium; warranty; porn; quality; vpn; high; instagram; guarantee
book; video; mastery; academy; market; figure; facebook; amazon; trade; dan
Fakedoc_defense (N = 1173) 3 id; driver; license; fake; licence; drive; quality; track; high; australia
psd; template; passport; hq; card; statement; utility; editable; bank; fully
passport; scan; card; utility; custom; bill; verification; usa; uk; selfie
Fakedoc_drugs (N = 78725) gram; quality; ship; free; mdma; cocaine; pure; high; 5 g; ketamine
quality; pill; mdma; high; top; xtc; mg; europe; dutch; import
ship; pill; mg; xanax; sale; usa; duplicate; 10 mg; bar; price
free; 5 g; uk; top; thc; indoor; sale; aaa; grade; haze
ship; free; thc; 1 g; new; premium; fast; cannabis; g; day
All_sauf_drugs (N = 15844) 8 hq; card; usa; cc; full; bank; fullz; fresh; scan; email
account; premium; warranty; hq; extra; market; cheap; bonus; month; access
account; premium; warranty; porn; lifetime; extra; bonus; video; movie; include
hq; psd; template; perfect; full; bank; id; scan; license; driver
hq; card; egift; pdf; gift; money; save; lot; checker; code
full; pro; pack; crack; complete; vpn; security; program; gold; adobe
database; record; hack; leacked; plaintext; million; dtabase; leak; voter; log
perfect; replica; shoe; high; quality; vuitton; louis; lv; gold; gucci